CURE: an efficient clustering algorithm for large databases
SIGMOD '98 Proceedings of the 1998 ACM SIGMOD international conference on Management of data
The b-chromatic number of a graph
Discrete Applied Mathematics
ACM Computing Surveys (CSUR)
Modern Information Retrieval
CLARANS: A Method for Clustering Objects for Spatial Data Mining
IEEE Transactions on Knowledge and Data Engineering
Clustering and its validation in a symbolic framework
Pattern Recognition Letters
A tutorial on spectral clustering
Statistics and Computing
A new greedy algorithm for improving b-coloring clustering
GbRPR'07 Proceedings of the 6th IAPR-TC-15 international conference on Graph-based representations in pattern recognition
A new clustering approach for symbolic data and its validation: application to the healthcare data
ISMIS'06 Proceedings of the 16th international conference on Foundations of Intelligent Systems
Some new indexes of cluster validity
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Performance evaluation of constraints in graph-based semi-supervised clustering
AMT'10 Proceedings of the 6th international conference on Active media technology
A graph model for mutual information based clustering
Journal of Intelligent Information Systems
Hi-index | 0.00 |
This paper proposes an approach toward improving re-coloring based clustering with graph b-coloring. Previous b-coloring based clustering algorithm did not consider the quality of clusters. Although a greedy re-coloring algorithm was proposed, it was still restrictive in terms of the explored search space due to its greedy and sequential re-coloring process. We aim at overcoming the limitations by enlarging the search space for re-coloring, while guaranteeing b-coloring properties. A best first re-coloring algorithm is proposed to realize nongreedy search for the admissible colors of vertices. A color exchange algorithm is proposed to remedy the problem in sequential re-coloring. These algorithms are orthogonal with respect to the re-colored vertices and thus can be utilized in conjunction. Preliminary evaluations are conducted over several benchmark datasets, and the results are encouraging.